An HIV-1 vaccine should induce antibodies able to neutralize the virus in vivo. Most HIV neutralizing activity in infected humans is directed at the gpl20 glycoprotein. To be effective in vivo, vaccine-induced anti- gpl20 antibodies must be able to counter those viruses most commonly transmitted from donor to recipient. These are usually of the non- syncytium-inducing (NSI) phenotype, their growth in vitro being restricted to primary CD4+ lymphocytes and macrophages. Unlike the more cytopathic SI viruses, NSI viruses are unable to enter permanent T-cell lines. Furthermore, primary viruses, whether SI or NSI, are relatively resistant to neutralization by sCD4, which we have shown to be due to a reduced rate of sCD4 binding to virions at 37 degrees C. We believe that primary viruses did not evolve to resist neutralization by sCD4, but they may have done so to resist antibodies directed at the CD4 binding site on gpl20, which are abundant in HIV-1+ sera. The envelopes of primary viruses, and especially NSI viruses, may differ in some fundamental way from those of T-cell line-adapted viruses such as IIIB or MN. Thus it is essential for vaccine design to understand the interactions of primary viruses with neutralizing antibodies. We propose to compare primary viruses and genetically related, sCD4-sensitive variants adapted to growth in T-cell lines for their sensitivity to neutralization by monoclonal and polyclonal antibodies. We will also study NSI and Sl viruses isolated sequentially from infected humans to establish whether a shift in viral phenotype alters neutralization sensitivity. For these experiments, viral envelopes will be inserted into the NL4-3 infectious molecular clone, creating chimeric viruses for genetic manipulation. We will explore whether. minor sequence variations affect the ability of gpl20 monomers to bind MAbs directed at, e.g., the CD4 binding site, or whether, as found with sCD4, reductions in neutralization sensitivity are due to a more subtle effect only seen with oligomeric glycoproteins on an intact virion. We will also test whether increased neutralization sensitivity is correlated with destabilisation of the gpl20-gp4l linkage. These studies will increase our understanding of the interactions of primary viruses with neutralizing antibodies, which may help to generate effective vaccines against HIV.